Q4 is potentially unfixable

[Oeloe]

Update in the ESR thread about the 1.1 beta:

*UPDATE* SyncError has posted the following message regarding hitboxes:

"Okay, kinda big breaking disappointing news of sorts. We just got done with some of the hitbox testing. Things do not look good. Reducing the hitboxes causes a slew of clipping errors both in multiplayer and singleplayer. In multiplayer it causes adverse effects on player collisions with the map and others -- in single player it pretty much just breaks most of the maps causing you to get places you shouldn't, and then getting stuck until you go to the console to quit or noclip yourself out.

Then there's the other whole issue. I went over this with serveral here, but Tim showed us the numbers and hitboxes for all the Quake games. Its literally the same box, size, positioning and with similiar relations to world collisions. The box isn't any larger than it was in Q3, we're all just better railers. Now reducing the box would make it more difficult to hit and would once again balance our high rail percentages out, but it can't be done without breaking much of the game. The conclusion in the end was that if people really are getting that much better at games than in future FPS they'll try to reduce the hitbox size, but on a game that's already been built around those numbers its causing too many errors to go back and change now.

In the end, everyone else here feels that no one has an advantage over another play and so since its neither a "bug" in their eyes or something that gives one player an advantage over another in a match it could only be fixed if it didn't bring on dozens of new bugs, which it currently is.

They are looking at other options, but while the hitbox reduction is what everyone is crying for and is a good fix in concept, it may not be possible based on all of these test results.

I'm trying to get all of the community's desires answered, but we can only make changes if they are for the best and while its definately harder to hit the smaller hitboxes, its incredibly ugly what it does to collisions. The reduction also decreased players jumping abilties (due to these collision changes) as we feared and trying to make up for the decrease by increasing a players jump height causes almost as many issues as the hitbox reduction itself.

This is not to say that there is no fix, we may just have to look at other options. "

It looks very much like this game is beyond redemption. The shape it's currently (hitboxes and player movement) in is probably the result of avoiding the buggy mess of the engine/game code.

On the other hand, the Q4W increases player movement speed a lot and rocket speed to (right?), which should make online hit detections extremely inconsistent according to SyncErrors claims. I haven't tested the mod online so i can't say anything about it, but perhaps someone else can.

Anyway, i've adjusted my hopes again. I have no idea what to expect but it looks bad.

[Runningman]

its fuct.

http://www.overdrivepc.com/forum/showthread.php?t=9678

[str3nd]

I really don't get the issue here. Making the scale of the entire hitbox smaller may break things(jump hight wise), but shouldn't making the hitbox more SLENDER -in MP only-avoid those issues?

[iluvquake4]

Excuse my ignorance, but why didn't they use per-poly/pixel hit detection like in Doom 3?

[DooMer]

That stuff really isn't even broken to begin with.

[Runningman]

I really don't get the issue here. Making the scale of the entire hitbox smaller may break things(jump hight wise), but shouldn't making the hitbox more SLENDER -in MP only-avoid those issues?

the hitbox and model size are pretty much a part of the baseline code in the games source. to change that would be like a domino effect, effecting everything in its path.

That stuff really isn't even broken to begin with.

your clueless,i geuss thats why your names doomer

[pjw]

Maybe it's just me, but I don't quite get how you went from this:

Reducing the hitboxes causes a slew of clipping errors both in multiplayer and singleplayer. In multiplayer it causes adverse effects on player collisions with the map and others -- in single player it pretty much just breaks most of the maps causing you to get places you shouldn't, and then getting stuck until you go to the console to quit or noclip yourself out.

plus this:

Its literally the same box, size, positioning and with similiar relations to world collisions. The box isn't any larger than it was in Q3, we're all just better railers.

to arrive at this:

It looks very much like this game is beyond redemption. The shape it's currently (hitboxes and player movement) in is probably the result of avoiding the buggy mess of the engine/game code.

Sorry, but it really kinda pegs my "dramatic horseshit" meter. (Oh, and the thread title? That helps too.)

When you take something as basic as the size of the player collision box (which has been the same throughout every Quake game), and change it in a game that was coded to work with the previous size...surprise! You get lots of errors and problems! Golly!

[Edit: Beaten by DooMer and Runningman...]

[DooMer]

I really don't get the issue here. Making the scale of the entire hitbox smaller may break things(jump hight wise), but shouldn't making the hitbox more SLENDER -in MP only-avoid those issues?

the hitbox and model size are pretty much a part of the baseline code in the games source. to change that would be like a domino effect, effecting everything in its path.

That stuff really isn't even broken to begin with.

your clueless,i geuss thats why your names doomer

The hitbox is the same size as q3. Is q3 broken too?

[spookmineer]

If the hitboxes are the same size, what could cause the increased rail accuracy in Q4?

[Mogul]

Excuse my ignorance, but why didn't they use per-poly/pixel hit detection like in Doom 3?

Takes up LOTS more bandwidth and also negates the fairness of using different player models. The irony is, Quake 4's models are so bland and the same anyway that they kind of defeated that purpose of having hitboxes. Sort of. It's fair, yes. And also fucking boring.

[DooMer]

I was really hoping that this would be the quake where the rail's damage was nerfed. At least a longer recharge. I whined about it during the beta, but I was shunned. I'm used to getting fucked with that thing, so it's no biggie.

[Tormentius]

That stuff really isn't even broken to begin with.

Agreed but people will always find something else to bitch about.

[iluvquake4]

I expect to see these kinds of anti-Quake 4 posts at Raven's forum, but here I thought the threads were kept a bit more optimistic. Some people will never like the game, so be it. Do they have to keep bashing it? Move on already ...

[iluvquake4]

The title of this thread misrepresents the developer's post.

The developer stated:

"This is not to say that there is no fix, we may just have to look at other options. "

[iluvquake4]

Thank you. :icon31:

[Runningman]

at least be professional about the hitBOX and make it the PLAYER and not a box....... seems very non-comp friendly. even if it where several smaller boxes stacked in the shape of the model like legos to save for bandwidth. this may even clear up the bs rail issue.
"oops my aims a lil off today, i shot over your shoulder...... oh you died anyway...cool"

Do they have to keep bashing it? Move on already ...

most comments are far from bashing, it was peoples expectations, a step forward in fps. q4 is not this.

[Goofos]

If the hitboxes are the same size, what could cause the increased rail accuracy in Q4?

Netcode? Wrong hit detection? Lag?... :>

[Runningman]

heres a quick visual

** EDIT ** msg for image **

[Mogul]

The complaining almost all is coming from hardcore players that want a game that makes SENSE. This isn't just a fun game for a good deal of us; it's more similar to a sport. And when the rules to the game/sport are broken, we want them fixed.

If something doesn't make sense in the weapon balance or hitbox or whatever, we're going to complain because we _know_ it can be done better, and we don't want a game that's obviously not in the polished shape that we left Q3 in.

That said, it takes time, and I think that in the end, between id actually listening to what the community wants the game to be (at a very base level; esr) and Q4Max, we're going to have a very, very good game.

It takes time, but every single complaint is warranted.

[Runningman]

yes riddla but anything rounded would lag everything, we are a ways off from the box being that detailed, unless you want a 100 ping in your own state.

The hitbox is the same size as q3. Is q3 broken too?

nope, but this isnt 1999 anymore.

[Lenard]

The hitbox is the same size as q3. Is q3 broken too?

nope, but this isnt 1999 anymore.

Doomer is so many times better than you. And it is not about what year it is, it is about what makes a better game.

[spookmineer]

Unless I need new glasses, Q4 hitbox is actually smaller then Q3 hitbox?

[iluvquake4]

According to the Q4 Max guys the hitboxes for Q4 are identical to Q3 and Q3 has the same exact issues. Why people are upset about Q4 hitboxes is a mystery.

[KMG]

models=polygons
hitbox=polygons

dont see why it would be hard to make a roughly upside-down bowling pin-shaped hitbox and it adversely affect ping, etc.

Could someone get full-on nerdy technical as to why/why not?

I am assuming that it is because there is simply tons more code attached to a hitbox. Not only is that hitbox firmly wedged into the heart of the netcode, there is just lots of math for the surface. What does the player model do? Get animated, lit, and drawn. Everything else that makes a player entitiy a player entity is attached to the hitbox. It moves and rotates, dragging the player model along with it. It detects collisions from tons of entities, walls, projectiles, and needs to be precise in the netcode. Nobody cares if the model is rotated a little bit the wrong way. Hell, in Q3 the model would often be entirely outside of the hitbox if you went off a jumpad. And it can't really be changed because the entire engine was built around this one size of hitbox. It's hard-coded. You can't change "HIT_BOX_WIDTH" to 50. It was not designed to accept dynamic hitbox sizes. All of this seems a little odd though, because weren't there monsters in SP that had entirely different hitboxes? Or is that completely unrelated due to the per-poly thing? Whatever.

[Oeloe]

@ Runningman: Syncerror said they tried using a second, smaller hitbox for hit detection inside the current one for (geometry) collision detection. It allegedly caused huge lags.

I really wonder what's actually causing these almost doubled rail accuracy percentages though. It obviously has nothing to do with people becoming better at railing because they don't get the same percentages in Q3. :icon29:

I'm more worried about the comments about even increasing rocket speed messing up hit detection. Hyperblaster bolts move much faster than rockets too don't they, so are their hits not detected properly/consistently?